# MOSS Voice-Acting — Manual & Studies Hub

A measured, audio-backed usage guide for `laion/moss-tts-local-transformer-4.55b-voice-acting-v2` (4.55B expressive voice-acting TTS). For any target it gives the best prompt, which LoRA at which merge %, the expected reward, and measured side-effects. This is the agent-readable mirror of the [HTML hub](https://laion-ai.github.io/moss-voiceacting-manual/site/index.html).

**Newest (Aug 2026) — if you want a vocal burst at a specific place:** put the tag inline in the SCRIPT *and* merge that class's burst adapter at **λ 0.75–1.0**. Measured over 1,440 generations: the prompt alone lands the burst **23.6 %** of the time, the emotion adapter **alone lowers that to 16.7 %** (below no-adapter), and the burst adapter at 1.0 raises it to **71.9 %**. λ 0.25 is within noise of no adapter. Keep any emotion adapter at **≤ half** the burst dose. Cost is **blend**, not genuineness. Beware the **truncation wall** — the model can produce the burst and stop; never put a silence beat straight after it, and verify the words *after* the burst survive into an ASR transcript. [Listen](https://projects.laion.ai/laion-moss-local-1.5-voice-acting-4.55b/burst_lora_with_without.html) · [full chapter](https://laion-ai.github.io/moss-voiceacting-manual/site/recipes/bursts_merging_evaluation.html)

**Also new — building a style LoRA:** on the sports-commentator task, training on **real recordings only** beat a synthetic+real mix in **11 of 12** matched configurations (paired Wilcoxon p = 0.0010), and transferred across a language boundary (German training → English prompts). Adding synthetic data of the target style *diluted* the register.

**From the burst/merge/evaluation run:** for a **vocal burst inside a sentence**, λ = 0.75 is the knee (91 % of full-merge presence at a much better blend) and `burst@1.0 + emotion@0.5` is the best cell of 2,304 generations — but a *half-merged* burst adapter paired with a half-merged emotion adapter is the **worst**, because the two compete. Cross-lingual reference audio (Japanese anime → DE/EN) does **not** transfer the voice — similarity **0.188** over 1,280 generations against a **0.105** unrelated-speaker floor — but pitch still tracks the reference (r = 0.591, compressed toward the middle), so it is a cheap source of **new** character voices rather than a cloning method. And: always evaluate against a **real-audio ceiling control**; a sports-commentator sweep looked like a win until real human commentary scored the same as the un-adapted base. See the [burst & evaluation page](https://laion-ai.github.io/moss-voiceacting-manual/site/recipes/bursts_merging_evaluation.html).

**New (Aug 2026):** reward is `× (1 − WER)` (stops over-acted, unintelligible takes); at high energy add stabilizers `vn_ARSH_high@0.5 + vn_BRGT_high@0.5` (+`vn_S_NARR_high` for storytelling) to keep speech clear; a cheap API brain (gpt-5.6-luna) matches the local Gemma-4-31B (`google/gemma-4-31B-it-qat-w4a16-ct`) at cents/task. Favourite narrator recipe: `vn_S_STRY_high@0.65 + vn_ARSH_high@0.20 + vn_BRGT_low@0.20` → S_STRY 0.998, ESTH 0.59, WER 0.0, ~11s.

## 🚀 Start here — conditioning manuals
*How to drive the model: for a target (emotion, voice dimension, delivery), what prompt + which LoRA at which merge % — with measured side-effects. Each has an agent-readable Markdown mirror.*

- **[🎭 Emotion conditioning manual (40 emotions)](https://laion-ai.github.io/moss-voiceacting-manual/site/emotions/index.html)** · [MD](https://github.com/LAION-AI/moss-voiceacting-manual/tree/main/md/emotions) — Per emotion: the best prompting strategy **with and without** the emotion LoRA, the best merge λ, expected reward, and the top-3 ± correlations on other emotions / VoiceNet dims / genuineness / blend / quality. Start here to make a voice *feel* something.
- **[🎚️ VoiceNet dimension manual (57 dims × high/low)](https://laion-ai.github.io/moss-voicenet-manual/)** · [MD](https://github.com/LAION-AI/moss-voicenet-manual/tree/main/md) — Per speech dimension (tempo, warmth, arousal, roughness…): the dose-response across 25→125% merge, the best dose, the cost to genuineness/blend/quality, and the ± side-effects. The reference for *how a voice sounds*.
- **[🔁 Reinterpretation guide (procedural, at scale)](https://github.com/LAION-AI/voice-acting-search-agent/blob/main/REINTERPRETATION.md)** · [MD](https://github.com/LAION-AI/voice-acting-search-agent/blob/main/REINTERPRETATION.md) — A deterministic rule to pick LoRAs + scales + prompt from a clip's scores **without an LLM call**, so you can reinterpret a huge corpus (e.g. 1M DramaBox clips) — including the LoRA hot-swap / batch-bucketing answer.
- **[🔥 Expressive & NSFW recipe book (agent warm-start)](https://laion-ai.github.io/moss-voiceacting-manual/site/recipes/expressive_nsfw.html)** · [MD](https://github.com/LAION-AI/moss-voiceacting-manual/blob/main/md/recipes/expressive_nsfw.md) — Validated, ready-to-use LoRA+prompt recipes distilled from autonomous agent runs — per emotional target (desire, ecstasy, teasing, dominance, anger, pain…) and the explicitness LoRAs. Written for a fast warm-start (human or agent). 18+, synthetic.
- **[🎭 Contained / masked emotion (holding a feeling back)](https://laion-ai.github.io/moss-voiceacting-manual/site/recipes/contained_emotion.html)** · [MD](https://github.com/LAION-AI/moss-voiceacting-manual/blob/main/md/recipes/contained_emotion.md) — **Updated (Aug 2026 — corrected by a paired-audio re-run).** How to make an *intense* emotion sound *held back* (cold controlled rage, forced-calm terror, swallowed grief) instead of merely *mild* — and how to **measure** it. Turning the emotion down FAILED (→ sounds weak) and VoiceNet tension does **not** separate contained from free; the mechanism is **emotional masking** — keep the emotion present and drop **vulnerability** (`vn_VULN_low`, light + prompt-led over a `char_genuine/human` scaffold) so the feeling stays HIGH while **VULN drops** below a matched free take. **The corrected headline: masking is not a general trick.** With real paired ΔVULN it **works** where the signal is effort/arousal/social-surface (**Fatigue**, **Embarrassment**, **Pain**, and the cognitive/low-arousal states **Emotional Numbness / Intoxication / Helplessness** — best at *intense*), and **reverses** where vulnerability *is* the emotion (**Sadness +1.29**, **Pride +1.09**, **Distress**, **Affection** read *more* exposed held-back → ship the free take); **Fear/Doubt** erase rather than mask. The old "Anger works / moderate contains, intense flattens" story was the four-anchor + sparse-unpaired pass and does **not** survive paired measurement. Includes the **decision rule up front**, the **{moderate,intense}×{free,suppressed}** 2×2, the **target-band fitness**, validated cues, and the full 40-emotion + deep-re-run tables.
- **[💥 Vocal bursts, merge doses & how to evaluate a LoRA](https://laion-ai.github.io/moss-voiceacting-manual/site/recipes/bursts_merging_evaluation.html)** · [MD](https://github.com/LAION-AI/moss-voiceacting-manual/blob/main/md/recipes/bursts_merging_evaluation.md) — **New (Aug 2026).** The measured dose curve for **vocal-burst adapters inside a sentence** (presence 27 %→71 % from λ 0.25→1.0, but blend 5.73→3.97 — **λ 0.75 is the knee**, and Sexual_Lust/Pain get *worse* at 1.0), how burst and emotion adapters **interact** (`burst@1.0 + emotion@0.5` is the best of 2,304 generations; a half-merged pair is the worst), cross-lingual reference audio as a source of **new character voices** (similarity 0.188 vs a 0.105 floor — not a clone; pitch tracks at r = 0.591 but compresses toward the model's register), and the evaluation checklist — real-audio ceiling control, multiple-comparison correction, and why validation loss has now failed to rank checkpoints three times.
- **[🗣️ Voice consistency across clips while the emotion changes](https://laion-ai.github.io/moss-voiceacting-manual/site/recipes/voice_consistency_continuation.html)** · [MD](https://github.com/LAION-AI/moss-voiceacting-manual/blob/main/md/recipes/voice_consistency_continuation.md) — **New (Aug 2026).** Self-contained recipe for keeping ONE character's voice across several clips while the emotion changes, matched to a reference recording. `mode="continuation"` with an **anchor + 4 s tail** prefix (chaining from the full previous clip collapses identity, 0.691 → 0.692 → **0.280**; anchor+tail gives 0.777 → 0.787 → **0.835**), identity **enforced** rather than ranked (regenerate < 0.82, VC < 0.75, reject < 0.68, sequence scored ½ mean + ½ worst), the four continuation API gotchas, how to make the emotion actually turn, the seed trap, and why none of it applies to screams. No acting-challenge context needed.
- **[🎭 Acting challenges: running an autonomous casting agent](https://laion-ai.github.io/moss-voiceacting-manual/site/recipes/acting_challenges.html)** · [MD](https://github.com/LAION-AI/moss-voiceacting-manual/blob/main/md/recipes/acting_challenges.md) — **New (Aug 2026).** The consolidated playbook from ~2,000 generations across nine casting challenges: continuation from an **anchor + 4 s tail** (chaining from the full previous part collapses identity, 0.691 → 0.692 → **0.280**), identity **enforced** rather than ranked (resample below 0.82, reject below 0.68, score speaker as **½ mean + ½ worst part** → 0.662 → **0.784**), why **seeds must vary per task** (all nine challenges drew seed 7110 and repeated the same burst — a listener caught it), **never open or close a part with a burst** (it lands on the seam), why the ranker was **selecting against burst blending** (weight 0.00 vs genuineness 0.34, correlation −0.21), the **narration index** for "it read the line instead of acting it", the **saturating tempo dial** (asking 4 gets 2.8), and that non-verbal parts are a **different regime** — every remaining identity failure is a scream.
- **[📚 Acting challenges — the FULL record (page tree)](https://laion-ai.github.io/moss-voiceacting-manual/site/recipes/acting/index.html)** · [MD](https://github.com/LAION-AI/moss-voiceacting-manual/tree/main/md/recipes/acting) — **New (Aug 2026).** The complete consolidated record behind the playbook above: [**the best recipe for each of the nine challenges**](https://laion-ai.github.io/moss-voiceacting-manual/site/recipes/acting/per_challenge.html) with full verbatim captions, adapters, doses and sampling; [version history](https://laion-ai.github.io/moss-voiceacting-manual/site/recipes/acting/versions.html) with every cross-generation number; [what worked](https://laion-ai.github.io/moss-voiceacting-manual/site/recipes/acting/what_worked.html) / [partly worked](https://laion-ai.github.io/moss-voiceacting-manual/site/recipes/acting/what_partly_worked.html) / [failed](https://laion-ai.github.io/moss-voiceacting-manual/site/recipes/acting/what_failed.html); [metrics](https://laion-ai.github.io/moss-voiceacting-manual/site/recipes/acting/metrics.html); [all 351 prompts verbatim](https://laion-ai.github.io/moss-voiceacting-manual/site/recipes/acting/prompts.html); and the [edge-case swarm](https://laion-ai.github.io/moss-voiceacting-manual/site/recipes/acting/swarm_edgecases.html). Headlines you will not find on the summary page: the 8-term casting score is effectively a **WER detector** ((1−WER) correlates +0.91…+0.98 with the final score, genuineness +0.00); **three of the eight terms are dead** (`min(emo_peak,1)` clipped in 92 % of assemblies, `arc` at exactly 1.0 in 84 %, `qual` an unnormalised near-constant worth 31 % of the score); the Gemini supervisor was **rating presentation order** — it picked the 4th or 5th take presented in 71 of 81 rounds and the agent's own top take in 3 of 108 — until the v7 rubric fixed it; and one whole run (`out_v5_real`) has its **results paired with the wrong plans** (every `res.json` is 2.2 h older than its `req.json`).
- **[🔤 Prompt notation — brackets, capitals, line length](https://laion-ai.github.io/moss-voiceacting-manual/site/recipes/prompt_notation.html)** · [MD](https://github.com/LAION-AI/moss-voiceacting-manual/blob/main/md/recipes/prompt_notation.md) — **New (Aug 2026).** The four notation rules, each one learned by breaking it. **Round** brackets for delivery cues *and* vocal bursts `(screams)`; **square** brackets for pauses `[pause]`; **never capitalise a tag** (`(SCREAM)` is read out letter by letter); and **≥10 words per line**, because at ~2.8 words/s a shorter line cannot fill three seconds. Includes the trap that makes this expensive: the WER reference strips `\(...\)`, so an angle-bracket tag stays in the reference and every burst-tagged take is charged word-error for not pronouncing "sobs" — the ranker selecting *against* the takes the tags were added to help. Plus the publishing bug where audio and scores came from different runs.
- **[🏭 Scaling the voice-profile corpus to 500 voices](https://laion-ai.github.io/moss-voiceacting-manual/site/recipes/scaling_to_500_voices.html)** · [MD](https://github.com/LAION-AI/moss-voiceacting-manual/blob/main/md/recipes/scaling_to_500_voices.md) — **New (Aug 2026).** Self-contained build note and cost appendix for going from one reference voice to **500**: the **832-condition** matrix per voice (320 emotion + 456 VoiceNet + 28 edge + 24 character + 2 sports + 2 explicitness), the **two subsets** that make it useful (**A** = 32 candidates each on a *different* sentence from that condition's pool, EN/DE parallel; **B** = 16 candidates on sentences *shared* across all sub-conditions of a parent, for controllability fine-tuning), **19,963,500 candidates**, **40,408 h** of audio, **3.02 TB**. Costed in **GPU-hours** off one complete measured build: **15,731 GPU-h** audio + **146 GPU-h** text = **1,143,120 core-h** at 72 core-h/GPU-h, **1.66 % of REFORMO**, 2.6 days on 64 nodes. Includes the **verified defect that blocks the run** — the currently published profile was built from a group list where **0 of 808 groups carry `cond_key`**, so every group took the self-contained early-return path and became **32 takes of ONE sentence** (VoiceNet: 114 distinct texts for 456 groups) — plus a **runnable pre-flight checklist** with expected outputs (all 832 groups carry `cond_key`, A pools exactly 32 / B pools exactly 16, EN/DE parallel at 99.961 %, all **233** adapter keys resolve offline), the text pipeline (90 % FineTranslations paraphrase / 10 % hard-for-TTS domains at 1 % each, spoken form, length buckets, burst quotas, the minors-and-sexual-content rule), the ranking formula in use, WebDataset/parquet formats with the full 76-column list, the **150 s** worker stagger, and an honest risk list (the 10-word floor is **not** enforced on generated text: 298 lines fail it).
- **[🏗️ Running the 500-voice build on a supercomputer](https://laion-ai.github.io/moss-voiceacting-manual/site/recipes/running_the_500_voice_build.html)** · [MD](https://github.com/LAION-AI/moss-voiceacting-manual/blob/main/md/recipes/running_the_500_voice_build.md) — **New (Aug 2026).** The execution note for the page above: what actually happened when the 500-voice plan was run on JUPITER. **One voice per Slurm array element** (not a slice — the unit of failure should equal the unit of re-queue; a bad node in an 8-voice slice orphans seven healthy voices), `0-499%160`, sized against the **12 h QOS wall** rather than the partition size. The **150 s worker stagger** and why it is not optional — four concurrent model loads off a parallel filesystem put every worker on the node into **uninterruptible sleep**, and a 45 s stagger cost a dead 45-minute job. **Three-layer resumability** (shard markers / group-level resume from the data / a five-way completeness gate that is the only writer of `VOICE.DONE`), including the trap that cost three runs — a *failed* group writes a row carrying its `gid` too, so an unconditional `done.add(gid)` turns any transient failure into a permanent hole no restart can fill — and why check 5 (**parquet key set == tar member set**) is not implied by the two counts. Measured **1,405 gen/GPU-h** and **28.65 GPU-h** in-process per voice, but **31.2 GPU-h billed** — report the billed number, the gap is 9 %. Plus **three silent defects found in the corpus that had already shipped**: **5,157–5,220 Title-Case burst tags per voice** that the model spells out letter by letter instead of performing (and that the WER stripper then charges word-error for), and a burst density realising **33.3 %** of a configured **0.50** because the placement step drops any burst with no interior position in *both* languages and logs nothing — a knob whose realised value comes from a lossy downstream step is a request, not a setting. And a third, which affects any corpus built through the same code path — the LoRA merge weight **decaying as λ^k** because the cache of "pristine" scalings is re-snapshotted, after dosing, every time the adapter LRU loads or evicts (λ = 1.0 is immune, eviction resets it, and nothing in any output shows it).
- **[✍️ Text-side recipes: make the WORDS carry the condition](https://laion-ai.github.io/moss-voiceacting-manual/site/recipes/text_carries_the_condition.html)** · [MD](https://github.com/LAION-AI/moss-voiceacting-manual/blob/main/md/recipes/text_carries_the_condition.md) — **New (Aug 2026).** 106-arm ablation. Disfluency ladder span 0.44 → 2.07 by writing the hesitation into the line; the explicitness dimension's adapter does **nothing** on its own (ladder span −0.006); chunking is a **negative result**; and merging a burst adapter into an edge case **halves speaker identity** for a marginal emotion gain — inline tags on laughter only, never the adapter.
- **[✍️ Writing the condition into the line](https://laion-ai.github.io/moss-voiceacting-manual/site/recipes/text_recipes_per_dimension.html)** · [MD](https://github.com/LAION-AI/moss-voiceacting-manual/blob/main/md/recipes/text_recipes_per_dimension.md) — **New (Aug 2026).** Which dimensions get better when you write the condition into the TEXT, and which get worse. Disfluency gains 4.7x span; chunking is actively harmed; the explicitness dimension is completely flat without it (span -0.006); edge cases gain intensity but lose blend and words. 106 arms, 1,696 generations.
- **[🎬 Multi-part performances & voice conversion](https://laion-ai.github.io/moss-voiceacting-manual/site/recipes/multipart_and_voice_conversion.html)** · [MD](https://github.com/LAION-AI/moss-voiceacting-manual/blob/main/md/recipes/multipart_and_voice_conversion.md) — **New (Aug 2026).** How to build a ~30 s performance with a dramatic arc out of 2–4 parts without it sounding like three different actors: **part 0 chooses the voice, every later part is generated with it as `reference=[...]`** (the argument `build_user_message` has always had and the first run never passed). Plus Chatterbox voice conversion as a *repair* when conditioning is not enough — chained, and kept only if ECAPA similarity rises and **DNSMOS** does not fall — crossfaded seams, ranking whole assemblies, and why a non-verbal part collapses to 0.3 s if you size `tokens` by word count.
- **[🏷️ The caption system: what a procedural caption means, and what just changed](https://laion-ai.github.io/moss-voiceacting-manual/site/recipes/caption_system.html)** · [MD](https://github.com/LAION-AI/moss-voiceacting-manual/blob/main/md/recipes/caption_system.md) — **New (Aug 2026).** How the procedural captions that describe voices are derived, and two defects repaired across **165,516,420 corpus rows** on 2026-08-23. The emotion clause used an **absolute** threshold across 40 heads that sit on different scales, so it named **Interest on 90.6 % of all captions** and Bitterness on 0.1 % — it was reporting the scale of the head, not the emotion of the clip; a top-10 %-per-emotion percentile gate over a 132,833,726-utterance ECDF takes Interest to **5.3 %**, makes all 40 emotions occur, and lets a clip honestly say **"no dominant emotion"** (17.88 % do). And **`GEND`/`BKGN` were rendered backwards** — high `GEND` is *masculine* (+0.842 with chest resonance; 89.2 % agreement against 379 design-spec genders) and high `BKGN` is *cleaner* (+0.786 with recording quality), so **83.1 M rows** changed the gender word and **79.5 M** changed the background word, ending captions that read *"good recording, very noisy background"*. The numeric columns were always right — only the prose inverted, so **if you trained on caption text you learned the opposite of the truth for those two attributes**. Includes what to do if you train on captions (the old string is preserved verbatim in `caption_general_v1`), why you must never flip the words yourself, and why the two captioners are not interchangeable.
- **[🔊 Edge-case delivery (screams, crying, laughter…)](https://laion-ai.github.io/moss-emotion-edgecase-evolution/)** — Reward-maximised recipes (prompt + λ + sampling) for hard non-speech-heavy deliveries — fear-scream, pain-scream/groan, sobbing, shivering, genuine laughter — evolved with and without the LoRA.

## 🔬 Interactive studies & audio grids
*Listen to the effects. Every clip is labelled with its recipe and scores.*

- **[⭐ Emotion × Voice conditions](https://laion-ai.github.io/emotion-voice-conditions/)** — 6 cloned voices × 40 emotions × 4 delivery conditions (intense/moderate × free/contained), best-of-32 with a speaker-similarity floor. The flagship study — shows that intensity, not containment, breaks a cloned voice.
- **[🧬 40-emotion evolution](https://laion-ai.github.io/moss-emotion-evolution-all/)** — Evolutionary search (10 generations, mean-of-8) for the best λ / prompt / sampling per emotion — the source of the per-emotion recipes used elsewhere.
- **[🎛️ Prompt × LoRA strength study](https://laion-ai.github.io/moss-emotion-prompt-lora-study/)** — Neutral vs prompt vs LoRA @50/100/150% for all 40 emotions, with a cross-steering heatmap — shows when the LoRA helps and when a good prompt alone wins.
- **[💥 Edge-case evolution](https://laion-ai.github.io/moss-emotion-edgecase-evolution/)** — The audio grid behind the edge-case recipes: screaming, crying, laughter, groaning, shivering, maximised with/without LoRA.
- **[🎚️ VoiceNet LoRA evolution (dose 0/50/100%)](https://laion-ai.github.io/moss-voicenet-lora-evolution/)** — Per-dimension controllability + best prompting at three doses — 50% already delivers most of the shift for most dimensions.
- **[↕️ VoiceNet high/low scaling grid](https://laion-ai.github.io/moss-voicenet-highlow-grid/)** — 57 dimensions × high/low × 5 merge doses (25→125%), 3 audio samples each — hear the full scaling sweep per dimension side by side.
- **[🗣️ Character-voice clusters](https://projects.laion.ai/character-voice-clusters/)** — VoiceNet-embedding clusters of the generated character voices with procedural captions — browse the space of distinct synthetic voices.
- **[♻️ Character LoRA refinement grids](https://projects.laion.ai/moss-voiceacting-demos/refine_v2.html)** — 120 character voices refined by SIDON self-distillation (generate → filter → restore → warm-start FT) — before/after quality per character.

## 🤖 Autonomous search agent
*An agent (local or API LLM brain) that *searches* for merge + prompt recipes on its own, guided by the manuals above and scored by the full perceptual stack.*

- **[📊 Results & model comparison](https://laion-ai.github.io/voice-acting-search-agent/)** — Single-dimension optimisation, the 4-brain comparison (gpt-5.6-luna vs Gemma-4 31B/12B/MoE, side by side with audio), the supervised edge-case swarm, and the corrected per-dimension runs.
- **[🏆 Which brain? (MODEL_COMPARISON.md)](https://github.com/LAION-AI/voice-acting-search-agent/blob/main/MODEL_COMPARISON.md)** · [MD](https://github.com/LAION-AI/voice-acting-search-agent/blob/main/MODEL_COMPARISON.md) — Gemini-judged, same-budget: gpt-5.6-luna (API) scored 9/9/10 — tied with the local Gemma-4 31B (9/10/9), ahead of 12B/MoE. Both independently found the stabilizer adapters.
- **[💸 Cost estimates (COSTS.md)](https://github.com/LAION-AI/voice-acting-search-agent/blob/main/COSTS.md)** · [MD](https://github.com/LAION-AI/voice-acting-search-agent/blob/main/COSTS.md) — Measured token/$ figures: an API brain runs a search for cents per task (luna: $0.31 for 4 missions, no GPU), and supervision is ~$0.007/verdict — see the 100/1000-task projections.
- **[🐝 Swarm design (SWARM_PLAN.md)](https://github.com/LAION-AI/voice-acting-search-agent/blob/main/SWARM_PLAN.md)** · [MD](https://github.com/LAION-AI/voice-acting-search-agent/blob/main/SWARM_PLAN.md) — The plan for scaling to a 32–64-agent self-improving swarm: two-tier hive-mind knowledge (raw experience vs. approved manual), task registry, supervisor loop, offline consolidation.
- **[🧠 What the agents have learned](https://github.com/LAION-AI/voice-acting-search-agent/blob/main/docs/experience-digest.md)** · [MD](https://github.com/LAION-AI/voice-acting-search-agent/blob/main/docs/experience-digest.md) — The distilled, validated learnings folded into every agent's context: winner recipes, the stabilizer adapters, the (1−WER) reward, and the task-difficulty map (what the TTS model can and cannot do).

## 📖 Taxonomies (definitions)
*The vocabularies everything is scored against.*

- **[VoiceNet taxonomy (base)](https://projects.laion.ai/Voice-Acting-Pipeline/voicenet_taxonomy.html)** — The 57 perceptual speech dimensions with 7-point anchor descriptions — what each VoiceNet score means.
- **[VoiceNet-Ext taxonomy](https://projects.laion.ai/Voice-Acting-Pipeline/voicenet_extension_taxonomy.html)** — The extended dimension set used by the predictor heads.
- **[Acting edge-cases / challenges](https://projects.laion.ai/Voice-Acting-Pipeline/path_ac_acting_challenge.html)** — 266 hard expressive-delivery situations, each expressed as VoiceNet/EmoNet/vocal-burst tokens — also mirrored (JSON + docs) in the [voice-taxonomies](https://github.com/LAION-AI/voice-taxonomies) repo.
- **[Situation taxonomy](https://projects.laion.ai/Voice-Acting-Pipeline/path_sit_situation.html)** — The scene/situation vocabulary behind the DramaBox reinterpretation prompts.

## 🤗 Models & adapters (Hugging Face)
*The checkpoints and LoRA collections referenced throughout.*

- **[Base model — moss-tts…voice-acting-v2](https://huggingface.co/laion/moss-tts-local-transformer-4.55b-voice-acting-v2)** — The 4.55B expressive voice-acting TTS model everything conditions on.
- **[Emotion LoRAs v3 (40)](https://huggingface.co/TTS-AGI/moss-emotion-loras-v3)** — One rank-64 LoRA per emotion (≤25% Got-Talent), the adapters behind the emotion manual.
- **[VoiceNet dimension LoRAs (114)](https://huggingface.co/laion/moss-voicenet-dimension-loras)** — 57 dimensions × high/low, the adapters behind the VoiceNet manual and the high/low grid.
- **[German Mediathek HQ LoRAs (3)](https://huggingface.co/laion/moss-mediathek-hq-lora)** — Rank 16/32/64 trained on 43,612 segments / 185 h of high-quality expressive **German public-broadcast** speech (emotion half at EmoNet > 2.5 and class-balanced, plus an equal quality half ranked by vocal-burst blend + genuineness). Take `r64_epoch3`. **German adapter** — English output runs long (10–22 s for a 13-word line vs 3.7–4.8 s for German). [🎧 7 emotions × DE/EN, with and without the emotion adapter](https://projects.laion.ai/laion-moss-local-1.5-voice-acting-4.55b/mediathek_hq_lora_emotions.html)
- **[Vocal-burst LoRAs (64)](https://huggingface.co/laion/vocal-burst-lora-adapters)** — One adapter per non-verbal burst (chuckle, gasp, sob, scream, moan, whistle…), each confirmed by a listening audit against real recordings and a no-LoRA baseline. Read the [merge-dose page](https://laion-ai.github.io/moss-voiceacting-manual/site/recipes/bursts_merging_evaluation.html) before choosing λ.
- **[Sports-commentator LoRAs (6)](https://huggingface.co/laion/moss-sports-commentator-lora)** — **Use `real_r64_e8`** — the default for energetic, real-sounding sports commentary, chosen by listening rather than by metric. Trained on **real German broadcast segments only**; the real-only run beat the mixed synthetic+real run in **11 of 12** matched configurations (paired Wilcoxon p = 0.0010). `real_r32_e2` is the low-WER/half-size alternative. Ships with its own negative result: on this task the base model already matches *real human* commentary on the judge metric, so the adapters buy consistency and intensity, not capability. [Side-by-side A/B](https://projects.laion.ai/laion-moss-local-1.5-voice-acting-4.55b/sports_lora_real_vs_mixed.html).
- **[Character LoRAs — refined (120)](https://huggingface.co/TTS-AGI/moss-character-loras-refined-public)** — 120 SIDON-refined from-scratch character voices (public mirror; a genuine-voices public repo also exists).